Issue with slatedb in infinite retry loop

[tve]

I'm copying lots files into a zerofs 1.0 filesystem backed by a hetzner bucket and it has gone into an infinite loop it seems. The following messages repeat over and over:

2026-01-21T22:31:05.575748Z  WARN slatedb::mem_table_flush: won't flush imm to l0 because
 too many l0 files [l0_len=16, l0_max_ssts=16]                                           
2026-01-21T22:31:05.992263Z  INFO slatedb::retrying_object_store: retrying object store o
peration [error=NotFound { path: "zeroborg-1.0-2b20/manifest/00000000000000010763.manifes
t", source: RetryError(RetryErrorImpl { method: GET, uri: Some(https://nbg1.your-objectst
orage.com/xxxxx/zeroborg-1.0-2b20/manifest/00000000000000010763.manifest), retries: 0,
 max_retries: 10, elapsed: 46.690825ms, retry_timeout: 180s, inner: Status { status: 404,
 body: Some("<?xml version=\"1.0\" encoding=\"UTF-8\"?><Error><Code>NoSuchKey</Code><Mess
age></Message><BucketName>borg-tve</BucketName><RequestId>tx000007272289effcad898-0069715
3a9-176fe121-nbg1-prod1-ceph3</RequestId><HostId>176fe121-nbg1-prod1-ceph3-nbg1</HostId><
/Error>") } }) }, duration=1s]                                                           
2026-01-21T22:31:06.573336Z  WARN slatedb::mem_table_flush: won't flush imm to l0 because
 too many l0 files [l0_len=16, l0_max_ssts=16]                                           
2026-01-21T22:31:06.998200Z  INFO slatedb::retrying_object_store: retrying object store o
peration [error=NotFound { path: "zeroborg-1.0-2b20/manifest/00000000000000010763.manifes
t", source: RetryError(RetryErrorImpl { method: GET, uri: Some(https://nbg1.your-objectst
orage.com/xxxxx/zeroborg-1.0-2b20/manifest/00000000000000010763.manifest), retries: 0,
 max_retries: 10, elapsed: 4.101836ms, retry_timeout: 180s, inner: Status { status: 404, 
body: Some("<?xml version=\"1.0\" encoding=\"UTF-8\"?><Error><Code>NoSuchKey</Code><Messaborg-tve
ge></Message><BucketName>borg-tve</BucketName><RequestId>tx0000003c1c7393ef606fa-00697153
aa-176e4180-nbg1-prod1-ceph3</RequestId><HostId>176e4180-nbg1-prod1-ceph3-nbg1</HostId></
Error>") } }) }, duration=1s]                                                            

The process didn't respond to ctrl-c, had to kill -9. Upon restart it did a bunch of compactions and now seems to work again.

[Barre]

Hi,

I opened an issue: #337